Temporal and Spatial Stability of Color Constancy Algorithms

Computational color constancy algorithms are commonly evaluated only through angular error analysis on annotated datasets of static images. The widespread use of videos in consumer devices motivated us to define a richer methodology for color constancy evaluation. To this extent, temporal and spatial stability are defined here to determine the degree of sensitivity of color constancy algorithms to variations in the scene that do not depend on the illuminant source, such as moving subjects or a moving camera. Our evaluation methodology is applied to compare several color constancy algorithms on stable sequences belonging to the Gray Ball and Burst Color Constancy video datasets. The stable sequences, identified using a general-purpose procedure, are made available for public download to encourage future research:

Download annotations and results

The file graybal_subsequences_list.txt defines our manual subdivision of the Gray Ball sequences into subsequences without video cuts.

The file grayball_subsequences_stability.xlsx defines our automatic subdivision of the Gray Ball subsequences into stable and unstable subsequences.

The remaining files report precomputed statistics for error and stability measures of the analyzed color constancy algorithms.



On the evaluation of temporal and spatial stability of color constancy algorithms
(Marco Buzzelli, Ilaria Erba) In J. Opt. Soc. Am. A, volume 38, number 9, pp. 1349-1356, OSA, 2021.

 author = {Buzzelli, Marco and Erba, Ilaria},
 year = {2021},
 month = {9},
 year = {2021},
 pages = {1349-1356},
 title = {On the evaluation of temporal and spatial stability of color constancy algorithms},
 volume = {38},
 number = {9},
 publisher = {OSA},
 journal = {J. Opt. Soc. Am. A},
 url = {http://www.osapublishing.org/josaa/abstract.cfm?URI=josaa-38-9-1349},
 pdf = {/download/buzzelli2021stability.pdf},
 doi = {10.1364/JOSAA.434860},
 projectref = {http://www.ivl.disco.unimib.it/activities/color-stability/}}